scholarly journals Klasifikasi Aktivitas Manusia menggunakan metode Ensemble Stacking berbasis Smartphone

2021 ◽  
Vol 1 (2) ◽  
pp. 53-58
Author(s):  
Firman Aziz
Keyword(s):  
Data Mining ◽  
Training Sample ◽  
Parameter Setting

Dengan perkembangan teknologi yang semakin pesat, kini smartphone dapat mengenali aktivitas manusia menggunakan sensor accelerometer dan gyroscope yang telah tertanam didalamnya dengan menghasilkan ratusan bahkan ribuan record dan membutuhkan metode data mining untuk melakukan pengelompokkan berdasarkan output tersebut. Metode data mining yang memiliki kinerja lebih baik dibandingkan metode lainnya adalah SVM tetapi sensitif terhadap parameter setting dan training sample yang menyebabkan performa tidak maksimal maka ensemble adalah solusinya. Penelitian ini mengusulkan penerapan metode ensemble Stacking untuk melakukan klasifikasi aktivitas manusia berbasis sensor accelerometer dan gyroscope. Hasil menunjukkan kinerja ensemble Stacking dengan akurasi 99.2%, sensitivity 99.6% dan specificity 98.7%.

Research on Outlier Mining Algorithm Oriented to Traffic Information

2011 ◽  
Vol 299-300 ◽  
pp. 1312-1315 ◽  
Author(s):  
Wen Ge Xie ◽  
Hai Hong Wang ◽  
Xin Cai Gu ◽  
Yan Feng Guo
Keyword(s):  
Data Mining ◽  
Distance Measure ◽  
Traffic Information ◽  
Manhattan Distance ◽  
Parameter Setting ◽  
Outlier Mining ◽  
Mining Algorithm ◽  
The Common

This paper designs a new outlier mining algorithm based on distance through introducing the "key attribute", which reduces the amount of data mining and increases the efficiency of outlier mining, simultaneously, improves the common distance measure and calculates with the improved Modifing Weighted Manhattan Distance(MWMD), the improved mining algorithm cancels the requirements of the parameter setting in the case of without affecting the mining results and gives the isolation degree of outliers.

scholarly journals IMPROVING PERFORMANCE OF INDUCTIVE MODELS THROUGH AN ALGORITHM AND SAMPLE COMBINATION STRATEGY

2001 ◽  
Vol 10 (04) ◽  
pp. 555-572
Author(s):  
HALEH VAFAIE ◽  
DEAN ABBOTT ◽  
MARK HUTCHINS ◽  
I. PHILIP MATKOVSKY
Keyword(s):  
Data Mining ◽  
Negative Impact ◽  
Predictive Performance ◽  
Training Sample ◽  
Model Combination ◽  
Combination Strategy ◽  
Combination Methods ◽  
Sample Data ◽  
Model Ensembles ◽  
Mining Model

Multiple approaches have been developed for improving predictive performance of a system by creating and combining various learned models. There are two main approaches to creating model ensembles. This first is to create a set of learned models by applying an algorithm repeatedly to different training sample data, the second applies various learning algorithms to the same sample data. The predictions of the models are then combined accordings to a voting scheme. This paper presents a method for combining models that were developed using numerous samples, modeling algorithms, and modelers and compares it with the alternate approaches. The presented results are based on findings from an ongoing operational data mining initiative with respect to selecting a model set that is best able to meet defined goals from among trained models. The operational goals to be attained in this initiative are to deploy data mining model(s) that maximizes specificity with minimal negative impact to sensitivity. The results of the model combination methods are evaluated with respect to sensitivity and false alarm rates and are then compared against other approaches.

Interval Set Representations of Clusters

2011 ◽  
pp. 659-663
Author(s):  
Pawan Lingras ◽  
Rui Yan ◽  
Mofreh Hogo ◽  
Chad West
Keyword(s):  
Data Mining ◽  
Supervised Learning ◽  
Information Age ◽  
Training Sample ◽  
Important Data ◽  
Amount Of Information ◽  
New Information ◽  
Set Representations ◽  
Classification And Clustering

The amount of information that is available in the new information age has made it necessary to consider various summarization techniques. Classification, clustering, and association are three important data-mining features. Association is concerned with finding the likelihood of co-occurrence of two different concepts. For example, the likelihood of a banana purchase given that a shopper has bought a cake. Classification and clustering both involve categorization of objects. Classification processes a previously known categorization of objects from a training sample so that it can be applied to other objects whose categorization is unknown. This process is called supervised learning. Clustering groups objects with similar characteristics. As opposed to classification, the grouping process in clustering is unsupervised. The actual categorization of objects, even for a sample, is unknown. Clustering is an important step in establishing object profiles.

Data Mining and Machine Learning

2020 ◽  
Author(s):  
Mohammed J. Zaki ◽  
Wagner Meira, Jr
Keyword(s):  
Machine Learning ◽  
Data Mining

Heart Rate Variability, Emotions, and Music

2010 ◽  
Vol 24 (2) ◽  
pp. 112-119 ◽  
Author(s):  
F. Riganello ◽  
A. Candelieri ◽  
M. Quintieri ◽  
G. Dolce
Keyword(s):  
Data Mining ◽  
Heart Rate ◽  
Heart Rate Variability ◽  
Vegetative State ◽  
Low Frequency ◽  
Emotional Reactions ◽  
Heart Beat ◽  
Healthy Controls ◽  
Frequency Spectra ◽  
Emotional Value

The purpose of the study was to identify significant changes in heart rate variability (an emerging descriptor of emotional conditions; HRV) concomitant to complex auditory stimuli with emotional value (music). In healthy controls, traumatic brain injured (TBI) patients, and subjects in the vegetative state (VS) the heart beat was continuously recorded while the subjects were passively listening to each of four music samples of different authorship. The heart rate (parametric and nonparametric) frequency spectra were computed and the spectra descriptors were processed by data-mining procedures. Data-mining sorted the nu_lf (normalized parameter unit of the spectrum low frequency range) as the significant descriptor by which the healthy controls, TBI patients, and VS subjects’ HRV responses to music could be clustered in classes matching those defined by the controls and TBI patients’ subjective reports. These findings promote the potential for HRV to reflect complex emotional stimuli and suggest that residual emotional reactions continue to occur in VS. HRV descriptors and data-mining appear applicable in brain function research in the absence of consciousness.

Post, Mine, and Be Disturbed: Social Media Data Mining

PsycCRITIQUES ◽  
2016 ◽  
Vol 61 (51) ◽  
Author(s):  
Daniel Keyes
Keyword(s):  
Data Mining ◽  
Social Media ◽  
Social Media Data ◽  
Media Data
Sign in / Sign up

Export Citation Format

Share Document